Heart rate analysis method, device and equipment

Abstract

According to the technical scheme, the invention relates to a heart rate analysis method, device and equipment, and the method comprises the steps of: acquiring waveform data of a heart rate with noise, and carrying out denoising preprocessing; creating a deep residual network, performing first sampling on the preprocessed waveform data, inputting the waveform data subjected to the first sampling into the deep residual network, and performing training to obtain a classification category and an error level of the heart rate; carrying out visualization processing on the sampled waveform data; taking the sampled waveform data as annotation data; adjusting the searching position of a millimeter radar wave, acquiring a plurality of heart rate measuring positions, and taking the position with the minimum heart rate error grade as the optimal monitoring position. The invention has the beneficial effects that the heart rate is accurately measured, so that a user can be guided to correctly use equipment; and data visualization is realized, so that the mode in the data is easier to find.

Description

technical field [0001] The invention relates to the field of computers, in particular to a heart rate analysis method, device and equipment. Background technique [0002] With the rapid development of sensor technology, signal processing chips, and artificial intelligence algorithms, the detection methods of vital sign parameters are becoming more and more extensive, such as smart watches to detect heart rate and blood oxygen, piezoelectric film belts to detect heart rate and respiration, millimeter wave radar to detect heart rate and respiration, etc. . These new vital sign parameter detection equipment are more convenient, comfortable and quicker than traditional oximeters, electrocardiographs, and monitors. [0003] The factors that affect the accuracy of vital sign parameter detection equipment can be attributed to four aspects: (1) the state of the human body, such as stationary, running, eating, sleeping, etc.; (2) the interaction between the human body and the sensor...

AI Technical Summary

Problems solved by technology

[0005] 1. Various usage scenarios have no constraints on the state of the human body. The result is strong interference. When a smart watch measures heart rate while running, there will be strong interference (relative movement between the wrist and the sensor). Piezoelectric film with millimeter-wave radar equipment In sleep monitoring, when calculating heart rate, the movement of limbs and the interference of breathing movement are dozens or even hundreds or thousands of times of useful signals

[0006] 2. There are uncertainties in the interaction between the human body and the sensor, the tightness of wearing a smart watch, the tightness of the contact position between the piezoelectric film sensor and the human body, the distance between the millimeter-wave radar and the human body, and the scanned body parts, etc. , these uncertainties result in varying degrees of signal-to-noise ratios, often very poor

[0007] 3. The signal processing hardware is immature, and the existing edge computing platforms are immature in terms of computing power, power consumption, and cost.

[0008] 4. The diversity of scenes and the uncertainty of the interaction between the human body and the sensor bring the complexity of data distribution, which determines the complexity of the algorithm and the number of samples required. Therefore, the new smart watch , piezoelectric film strips, millimeter-wave radar and other vital sign detection equipment require more complex algorithms and more samples

Based on these reasons, the detection results of the existing new vital sign parameter equipment are sometimes good and sometimes bad. The accuracy of the measurement results is high when the interference of the human body is quiet and the environment is small, and the measurement accuracy is low when the interference around the human body is large. These problems make the user very uncomfortable. Difficult to use measurement results

Images